Datasets:
File size: 1,651 Bytes
c83afe9 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 | import csv
import os
from PIL import Image
SIZE = (512, 512)
MASK_TYPES = ["inpainting", "outpainting", "random"]
def apply_mask(image1_path, mask_path, size=SIZE):
"""Return GT image: keep image1 pixels where mask=1 (white), black where mask=0."""
img = Image.open(image1_path).convert("RGB").resize(size)
mask = Image.open(mask_path).convert("L").resize(size)
black = Image.new("RGB", size, (0, 0, 0))
return Image.composite(img, black, mask)
def generate_gt_images(csv_file, output_dir):
if not os.path.exists(csv_file):
print(f"Warning: {csv_file} not found, skipping.")
return
with open(csv_file, "r", encoding="utf-8") as f:
rows = list(csv.DictReader(f))
os.makedirs(output_dir, exist_ok=True)
for row in rows:
id_val = row["id"].strip()
image1_path = row.get("image1_path", "").strip()
image2_path = row.get("image2_path", "").strip()
DATA_ROOT = "data"
image1_path = os.path.join(DATA_ROOT, image1_path)
image2_path = os.path.join(DATA_ROOT, image2_path)
img_path = os.path.join(output_dir, f"id_{id_val}.png")
if image1_path and os.path.exists(image1_path):
gt = apply_mask(image1_path, image2_path)
gt.save(img_path)
else:
print(f"Skip id={id_val}: image1_path missing or not found ({image1_path!r})")
print(f"Done: {output_dir}")
if __name__ == "__main__":
for mask_type in MASK_TYPES:
generate_gt_images(
f"metadata/Task_Image_Mask_{mask_type}_metadata.csv",
output_dir=f"data/Task_Image_Mask/{mask_type}"
)
|